[En-Nut-Discussion] CRC32 Lib
morty at gmx.net
Thu Jun 26 12:21:50 CEST 2008
Alain M. schrieb:
> Moritz Struebe escreveu:
>>> A slow but small variant could also be usefull for some embedded,
> What I mean is another implementation completely: one that does not use
> a lookup table, and very little memory. Very often in embeddeds space is
> more important then speed.
Did you look at the code? Both solutions are implemented. :-)
>>> both block or incremental.
> Block is just what you did. The whole bloc of data has to fit in memory
> at once and cpu will be used for as long as it takes.
Are we talking about the same code?
> Incremental is another aproach, where each byte is porcessed at a time,
> eg. as it is being transmitted. It has two advantages: no need to have
> memory for all the data and cpu is not blocked for so much time.
Yea, that's exactly what my code does. It's no big deal wrapping the
byte-wise solution in to a block-solution.
> Of course these are other implementations, but you might just have info
> about them somewhere, as you just researched the subject ;)
No didn't find anything suitable to my needs, making the lookuptable
optional and doing byte wise crc-calculations. - It's not like I can
load ~90K into ram to validate my Flash.
More information about the En-Nut-Discussion